The community member is trying to use llamaparse in an IngestionPipeline but is unable to get both Nodes and IndexNodes returned by the pipeline. The comments suggest that using the markdown element node parser may not work, as the ingestion pipeline assumes a single list of input/output objects. A community member suggests that a custom transformation class could handle it, but it would be "janky".
I am trying to use llamaparse in an IngestionPipeline but can't get to have both Nodes and IndexNodes returned by the pipeline. How can I do that? Thanks!
